Most American youth do not meet the Physical Activity Guidelines for Americans, and a major public health effort is underway to promote physical activity in young people. Monitoring this effort requires a system that provides valid and detailed information on physical activity behavior. Although accelerometry can be used in large-scale surveillance studies, it is expensive, burdensome and provides no information about specific forms or contexts of physical activity. Self-report instruments are practical for use in surveillance studies, but no current self-report instrument for youth provides both excellent validity versus an objective measure of physical activity and detailed information on specific forms and contexts of activity. The purpose of this study is to develop, pilot test, and validate a physical activity self-report instrument that has strong construct validity, matches well with the cognitive abilities of youth, associates strongly with objectively-measured physical activity, and contains items that will provide information about specific forms and physical/social contexts of physical activity in youth. The proposed study has two Aims: Aim 1 is to develop a physical activity self- report instrument for youth that has robust construct validity. We will review the literature on existing PA instruments and conduct focus groups with 96 middle school boys and girls to identify common types, frequency, and intensity of activities. From the data, we will identify a pool of candidate dichotomous (Yes/No) items and administer the resulting instrument to 500 middle school students. Item Response Theory will be utilized to select the items that best represent the construct of physical activity in 11-15 year-old boys and girls. Aim 2 is to examine the validity of the self-report instrument versus an objective measure of physical activity in youth. An independent sample of 300 students will complete two forms of the final self-report instrument after wearing an accelerometer for 7 days. Half of the sample will complete the instrument by providing only Yes/No responses. The other half will provide both Yes/No and frequency of participation responses. Analyses will test the validity of the new instruments versus daily minutes of moderate-to-vigorous physical activity measured via accelerometry. The proposed study carries high public health significance because it will develop a measure of physical activity in youth that has excellent validity and is practical for use in surveillance systems and large sample investigations. The study is innovative because it will be the first time a physical activity self-report instrument for youth is built with a proven program of test validation using state-of-the-art Item Response Theory. The information provided by the instrument will contribute importantly to the development and evaluation of public health policies and programs aimed at increasing physical activity in U.S. youth.